ECON majors with some background in linear algebra, functional programming, and statistics may be interested in this new physics course being offered in Spring 2018:
PHYS
476: Applied Machine Learning
Wednesdays,
4:00pm - 7:00pm
Room: PHYS 4221
Room: PHYS 4221
Credits:
3
This one semester course
introduces machine learning techniques that are becoming pertinent in the technology
industry. The course will focus on deep learning using a hands-on approach and
popular high-level libraries
(TensorFlow, Gensim, Keras, etc), and is designed for a broad audience of
intermediate students in related disciplines (any CMNS, Economics, linguistics,
etc.) in the sciences. It's goal is to give students an
understanding of the field and its capabilities, as well the tools to learn the
necessary extensions of the topic to apply it to their research.
Lectures will include introductions to Python and Linux, GPU
acceleration, cloud computing, neural nets, deep learning, natural language
processing, imagine recognition/computer vision, and AI safety.
Students are expected to have some background in functional
programming, linear algebra, calculus, and mathematical modeling. Some
proficiency in Python is strongly suggested. The course will be taught using a
combined lecture/laboratory approach, with coding exercises occurring
periodically to build basic proficiency with the techniques discussed in an
informal group environment.
For
more information on the course, contact the instructors: Matt Severson (stizashell@gmail.com)
and Justin Terry (justinkterry@gmail.com).
To register for the course, send an email request to ugrad@physics.umd.edu.
